"How do you perceive the time and space of a particular event? A hate crime, for example? Is it a local event? Could it be tied to something much broader—on a global scale?" Colin Flint asks.
Consider these events from 1998, sampled from a database kept by the Pennsylvania Human Relations Commission: 13 crosses burned in a backyard in Jefferson County; hate mail sent to a rabbi in York County; a swastika scratched into a car in Northampton County. The events may seem isolated. Yet Flint, assistant professor of geography, sees patterns in the data. He uses spatial statistics to model the geography of hate activity.
Flint chose to focus on Pennsylvania because of what he calls a "historical accident." A dispute within a Pennsylvania chapter of the Ku Klux Klan landed boxes of Klan records at the State Police headquarters. The Klan, which started in 1915 as a fraternal organization, became successful through pro-American, anti-Catholic, anti-Immigrant, anti-Semitic beliefs. Flint used the Klan data, and current data on hate activity collected by the Pennsylvania Human Relation Commission, to compare hate activity in the 1920s to hate activity in the 1990s.
Flint's first model, using data from the 1920s, is based on the number of Klan cells—local chapters—per 100,000 people, not the number of members or crime events, because Flint doesn't have that information. When mapped by county, the data show regions of relatively high or low numbers of cells throughout the state. Flint used a statistical technique called spatial auto-correlation to create the map: counties with high values surrounded by counties with high values are dark-colored; counties with relatively low values surrounded by low-valued counties are light; and counties with values that are not significantly high nor low are given an intermediate color. "It's not just one county, it's part of a regional cluster," says Flint. On Flint's map, rural regions of Pennsylvania, such as counties in the Northern tier, are dark because of relatively high numbers of cells.
"I wasn't surprised about the rural areas," says Flint. "What struck me is that it is so low in the suburban areas.
"Other research has shown the Klan to be strong in areas where there have been labor tensions based on ethnic divides, such as Pittsburgh and the anthracite coal mining regions," says Flint. "Those areas don't pop up here.
Flint suggests that high-valued clusters are in rural areas because of major social changes that were taking place. "There was fear of the modern in the 1920s," says Flint. "The country was undergoing a transition from rural to urban."
"People who join hate groups or commit hate crimes are perceiving changes in society which they think threaten them," he explains.
After its peak membership in the 1920s, the Klan became a smaller, more fragmented group with a dwindling membership. "They don't dominate the nativist movement today," says Flint. The Southern Poverty Law Center reports that today there are only 5,000 Klan members nationwide, while 17,000 people belong to almost 500 other hate groups.
"Hate groups today are very different, certainly not as organized or structured as the Ku Klux Klan," says Flint. "There is a lot of interaction though, among the people at the top of these groups," says Flint. "They're not in opposition. I would say they have different philosophies but that their meta-philosophy is compatible. The idea is that America is under attack from a global threat and needs to defend itself."
Flint modeled hate activity in Pennsylvania from 1984 to 1999 using the same statistical techniques that he used for the 1920s model. This time, he used a database of "tension incidents" that were reported to the Pennsylvania Human Relations Committee though local police, school districts, the American Civil Liberties Union, and local newspapers. "The Human Relations Commission's definition of 'tension incident' is broad enough to include actual hate crimes, flyers left on cars, demonstrations," says Flint.
When Flint mapped the data, a pattern just "popped out" at him—high concentrations of activity in the suburban areas of Philadelphia, Harrisburg, and York. He saw almost the reverse of his 1920s map: Instead of hate activity being clustered in rural areas, it had shifted to suburban areas.
"I'd seen this in some exploratory research," says Flint. A few years ago, while on the faculty at Georgia Tech, Flint mapped hate activity in the Atlanta area. "It showed almost a donut of activity around the city, in the suburban areas.
"Contemporary hate activity," Flint concludes. "Takes place in the suburbs."
So who commits hate crimes? Is it "blue collar, economically distressed white guys? Or, middle-class people whose dreams are no longer attainable?" Flint asks. "Academics are quick to assume that it's working class or undereducated people." But, that may not be the case. "It could be a younger generation," says Flint, the children of people who moved to the suburbs to pursue the American dream—a dream suburbia "perhaps failed to deliver," he adds.
"Now that we've targeted the places we need to do more research to figure out what's going on within them," says Flint. "It's time to get away from the numbers and talk to people."
Colin Flint, Ph.D., is assistant professor of geography in the College of Earth and Mineral Sciences, 302 Walker Building, University Park, PA 16802; 814-865-3433; flint@geog.psu.edu.